There are approximately 100,000 upper-limb amputees in the U.S. Regardless of amputation level, most of them have in common the need for a prehensor because it is through the hands that all individuals interact with the physical world. The RFA specifically enumerated "prosthetic...device design development for different functions..." and "methods of powering...mechanical powered devices." This research directly address both of these priority areas. The central goal of this proposed research is to improve prosthetic prehension for both body-powered and externally-powered prehensors. Specific projects include: 1. Vector prehensor - a voluntary opening prehensor with easily adjustable grip force which can be adjusted to the demands of the task, improving the efficiency of grasping and reducing the mechanical energy demands. 2. Variable mechanical advantage prehensor - a voluntary-closing device with enhanced gripping efficiency; i.e. rapid sizing with minimal cable excursion coupled with large grip force generation. 3. General prehension research - improvements to prehension applicable to any type of prehensor, including anthropomorphic fingers with nonlinearly compliant structure and variable hardness finger materials. 4. Quantification of prehensor performance - methods to quantify grasping performance in the laboratory, including the degree of force and torque that can be effectively applied by a prehensor. Proposed research includes the design, fabrication and laboratory testing of prototype prehensors. In addition, amputee testing will be performed, including subjective evaluation of activities of daily living, and quantitative measurement of grasping capability and 3D trajectories during grasping functions. Emphasis will be placed on practical improvements to prosthetic devices that have a high probability of reaching the end users, rather than remaining in the lab.